DE102012223662B4 - A method of generating a high frequency power and power supply system with a power converter for supplying a load with power - Google Patents

Abstract

A method for generating a high-frequency power that can be supplied to a load by generating an analog signal by a DAC (41) and amplifying it in an amplifier path (31-36), wherein the amplitude of the analog signal is modulated, characterized in that the amplitude of the analog signal is modulated by supplying to a digital signal input of the DAC (41) a sequence of digital values from which the DAC (41) generates the analog signal and generates the sequence of digital values by a signal data value stored in a signal data memory (61) is multiplied by an amplitude data stored in an amplitude data memory (62).

Description

The invention relates to a method for generating a high-frequency power that can be supplied to a load by an analog signal generated by a digital-to-analog converter (DAC) and amplified in an amplifier path, according to the preamble of claim 1.

The invention further relates to a power supply system having a power converter connectable to the load for supplying a load with power, the power converter having a first amplifier path according to the preamble of claim 6.

The DE 603 16 022 T2 describes a method and apparatus for optimizing the efficiency of a transmitter. A transmitting device in a wireless mobile device comprises an encoder and modulator which generates from an input signal a coded and modulated transmission signal and a data indicator signal according to the type of the coded and modulated transmission signal. A digital-to-analog converter is coupled to the encoder and modulator and generates an analog representation signal of the encoded and modulated transmit signal.

From Wikipedia: lookup tables it is known that lookup tables are used in computer science and digital technology to statically define information and use it at runtime of the program - to avoid time-consuming calculations or high memory consumption.

The WO 2008/150136 A1 deals with an arc detection device and a method for arc monitoring.

The US 2005/0237035 A1 describes a control system for a power supply.

Power supply systems, in particular systems that generate power at frequencies> 1 MHz, are used for example for laser excitation, in plasma coating systems or for induction applications. In such power supply systems, multiple amplifiers are often used to generate a total power from the power supply system. The power signals generated by the individual amplifiers are often not phase synchronous. Also, the power signals may have different amplitudes. The output power of each amplifier, d. H. the power signals often have to be combined into a total power, e.g. B. by combiners (combiner), or directly to a load, eg. B. a plasma electrode or a gas laser electrode. For combining the output powers, fixed phase relationships of the output powers are often required. In addition, the output power of the amplifiers should be matched in amplitude to each other.

A further problem is that it is difficult in such systems to adjust the power and in particular the degree of ionization in the load.

It is therefore an object of the present invention to provide a method and a power supply system with which the power in the load, in particular a degree of ionization of the plasma, can be adjusted.

This object is achieved according to the invention by a method for generating a high-frequency power, with the features of claim 1. As a load is in particular a plasma process or a gas laser process in question. By the amplitude modulation of the analog signal, the power supplied to the load can be controlled, and thus, for example, the degree of ionization of a plasma in a plasma load can be set and controlled in a particularly simple manner. Plasma load may be a processing plasma for etching or coating, or also a gas laser exciting plasma. The amplitude of the analog signal is modulated by supplying to a digital signal input of the DAC a series of digital values from which the DAC generates the analog signal. In this way, an analog signal can be generated particularly quickly and accurately.

The sequence of digital values is generated by multiplying a signal data value stored in a signal data memory by an amplitude data value stored in an amplitude data memory. In particular, the sequence of digital values can be generated by reading, with the aid of a counter, a sequence of signal data stored in a signal data memory, applying this sequence of signal data values to a multiplier, and multiplying this multiplier by an amplitude data value read from an amplitude data memory. In this case, an analog signal can be an analog voltage. Thus, a digital signal can be generated in a simple manner, which has information regarding the signal shape and the amplitude of the analog signal to be generated in the DAC. This has particular advantages when several analog signals are generated in parallel, which are then combined by a combiner, so that the signals to be combined can be adapted particularly easily and quickly to each other. Amplifier paths driven by an analog signal generated in this way are especially good for one Parallel operation of several amplifier paths suitable. The power generated in this way can easily be combined at the output of the amplifier paths. This results in a very fast and exactly tunable overall performance of a power converter.

The amplitude of the analog signal can be modulated by changing the amplitude data stored in the amplitude data memory. So the amplitude modulation can be done very fast. In addition, multiple amplifier paths can be changed so very fast in their output power and at the same time the synchronization of the amplifier paths with each other remains unaffected.

The amplitude of the analog signal can be modulated by changing the signal data stored in the signal data memory. Even so, the amplitude modulation can be done very quickly. In addition, several amplifier paths can be changed very quickly in their output power and a non-linear amplification and phase shift at different amplitudes of the analog signal of the amplifier can be taken into account.

A reference signal input of the DAC can be supplied to influence the analog signal, a reference signal. In particular, this allows the amplitude of the analog signal to be modulated. Thus, the analog signal can be influenced not only by the sequence of digital values but also by a reference signal supplied to the reference signal input. There are thus several options open to influence the analog signal and adjust it exactly, in particular to modulate its amplitude.

Due to the sequence of digital values, a digital ramp function can be realized. A digital ramp function is a slope given by digital values and a setpoint. For example, it can be specified that the power is increased to 1 kW with a slope of 100 W / ms.

Furthermore, it is possible to pulse between two amplitudes. This is particularly easy because the DAC only information about amplitude phase and frequency must be supplied. The realization of pulsing with two amplitudes is therefore particularly simple and very fast. Thus, for example, in a simmer operation, the discharge can be maintained at very low power without continuing the plasma process processing the material. Thus, a high degree of ionization is much faster by increasing performance than if you had to re-ignite.

The analog signal can be filtered. As a result, a particularly low-harmonic signal at the input of the amplifier path can be entered. This increases the efficiency of the overall system.

It can be generated by several DAC each an analog signal and amplified in an associated amplifier path. Thus, relatively low powers can be generated in parallel amplifier paths, which can then be combined to form a total power. For the individual amplifier paths inexpensive components can be used because of the relatively low power generated.

The output signals of the amplifier paths can be combined with a combiner to a total power. Thus, a high output power of the power converter can be realized in a simple manner.

The signal data memories and amplitude data memories can be described by a higher-level memory. The content of the memory allocated to the DAC can thus be determined and influenced by the higher-level memory.

The DAC can be controlled taking into account the signal propagation times and gain factors in the individual amplifier paths. Thus, a particularly accurate overall performance can be adjusted.

The total generated power and / or a power reflected from the load can be detected and fed to a higher-level controller. Thus, a particularly accurate control and control of the overall performance is possible.

The scope of the invention also includes a power supply system having the features of claim 6. In particular, a particularly fast and flexible amplitude modulation is possible, which can be used in plasma and laser applications. With the power supply system according to the invention, the amplitude modulation of a power converter with any envelope for controlling, for example, the degree of ionization of the plasma is possible.

The DAC is preceded by a logic circuit unit for generating a digital signal supplied to the DAC, wherein the logic circuit unit stores a signal data memory in which signal data values for generating an analog waveform are stored, an amplitude data memory in which amplitude data values for influencing the amplitudes of the analog signals are stored Multiplier for multiplying the signal data values by the amplitude data values.

Thus, a digital signal can be generated in a simple manner, which has information regarding the signal shape and the amplitude of the analog signal to be generated in the DAC. This results in the advantages mentioned with regard to the method.

The signal data values may in particular comprise phase information and / or frequency information for the analog signal to be generated.

At least two amplifier paths can be provided, each amplifier path being supplied with an analog signal generated by a digital signal from a DAC, and a logic circuit unit being connected upstream of each DAC. This measure results in a particularly compact design for high output power. This can save manufacturing costs and costs. In addition, such a structure becomes more reliable. Such a power converter is particularly easy to calibrate and tune.

A digital logic circuit connected to the logic circuit unit (s) may be provided. By the logic circuit unit, the data values to be multiplied can be selected.

The digital logic circuit and one or more logic circuit units may be integrated in a logic device. This results in a high degree of integration. It must be used few components, which saves space and is inexpensive.

The amplitude modulated analog signal may be a voltage, a current or a power. The term "amplitude" of a signal here means the maximum deflection of a periodically alternating variable.

The signal data values and / or the amplitude data values can be stored in look-up tables (LUT) and are particularly easy to select for a signal shape to be set and amplitude or amplitude modulation.

The signal and the amplitude data memory can be designed as look-up tables (LUT). This saves space since only a short code can be kept in the actual data files (with a high number of entries) and the associated long description can be used from the table. In the look-up table, precomputed results or other information can be defined for certain constellations. The calculation of values can thus be dispensed with, so that the analog signals and thus the power can be adapted extremely quickly.

The DAC may have a reference signal input. This reference signal input can be used to influence the generation of the analog signal. A change of the analog signal can be brought about very quickly in this way.

The reference signal input may be preceded by a drive circuit. With the aid of such a drive circuit, the input signal of the reference signal input can be changed rapidly, thereby also quickly causing a change in the generated analog signal, in particular its amplitude modulation.

A particularly compact construction results when the DAC and the associated logic circuit units are integrated in a direct digital synthesis (DDS module).

If a filter arrangement for filtering the analog signal is provided between the DAC and the amplifier path, a particularly low-harmonic signal can be input at the input of the amplifier path. This reduces annoying harmonics in the amplifier path. This increases the efficiency of the overall system.

The power converter can have more than two amplifier paths, to each of which a DAC is assigned, which supplies an analog signal to the respective amplifier path. This measure results in a particularly compact design for high output power. Such a power converter is particularly well tuned.

The amplifier paths may be combined with a combiner to combine the power generated in the amplifier paths into a total power. In this case, relatively simple combiners can be used, which are cheaper to produce, but place high demands on the phase position, amplitudes and frequencies of the input signals. The power converter can thereby be constructed in a particularly compact design and cost.

The combiner for combining the power generated in the amplifier paths can be designed without compensation impedances for input signals of unequal strength and / or phase. Such a combiner is particularly energy-efficient and cost-effective. In addition, such a Combiner very compact and can be built with few components.

The amplifier paths may comprise transistors in LDMOS technology. LDMOS stands for "laterally diffused metal oxide semiconductor". These are MOSFETs, which have so far been used mainly in the GHz range. The use of these transistors in the range below 200 MHz is not yet known. When used in amplifiers to produce a power that can be supplied to a plasma process or a laser, it has surprisingly been found that these transistors behave in LDMOS technology much more reliable than comparable conventional MOSFETs. This can be due to a much higher current carrying capacity. In particular, in experiments with multiple amplifier paths and at frequencies of 3.4 MHz, 13 MHz, 27 MHz, 40 MHz and 162 MHz, these types of transistors have shown a particularly high reliability. Another advantage of these transistor types compared to conventional MOSFETs is that the identical transistors can be used for the mentioned frequencies (3.4 MHz, 13 MHz, 27 MHz, 40 MHz and 162 MHz). This makes it possible to build amplifier and power supply systems which can be used at frequencies over several decades in the range from 1 MHz to 200 MHz, with a very similar or identical topology. These are frequencies that are often used in plasma processes and for gas laser excitation. The frequency can be set by simply changing the control of the digital-to-analog converter, the amplitude by changing the values in the digital memory or in the look-up tables or via the reference signal input of the digital-to-analog converter. Conventional MOSFETs at these frequencies often have problems with plasma processes when too much power supplied to the plasma process is reflected. Therefore, the power generated often had to be limited so as not to increase the reflected power beyond a critical limit. Thus, the plasma processes could not always ignite safely or operate in the desired power range. In addition, sophisticated variable impedance matching circuits have been provided to overcome these disadvantages. LDMOS transistors can now be used particularly advantageously if reflected power is to be expected to a considerable extent, as is the case, for example, in the supply of plasma processes. In connection with the above-mentioned combiner, the advantage of the LDMOS transistors is that much higher reflected powers can be absorbed by the transistors. As a result, the demands on additional impedance matching networks connected between the power supply system and the load become lower, and component and control costs can be saved in these impedance matching networks.

Each amplifier path may be assigned its own DAC with associated logic circuit unit, a higher-level memory, in particular a look-up table, being provided, which is connected to the DAC-associated logic circuit units. The higher-level memory can be used to describe the memory allocated to the DAC. This results in an overall system which is suitable for very high powers and which can be realized in a compact design. Also, the system is very easy to set and tune in terms of output power.

The higher-level memory can be integrated in the logic circuit. Furthermore, the drive circuit for the reference signal input of at least one DAC can be integrated in the logic component, which can be designed as a programmable logic device (PLD), in particular a field programmable gate array (FPGA). The logic circuit may be driven by a digital signal processor.

In particular, the DAC can additionally be integrated in this PLD or FPGA. Furthermore, the drive circuit for the reference signal input of at least one DAC can be integrated in a PLD or FPGA. The PLD or FPGA can be controlled by a digital processor, in particular by a digital signal processor (DSP). These measures contribute to the fact that a compact design for high power can be realized and that the power converter is very well tuned.

A measuring device, in particular a directional coupler, can be provided which is connected to a system controller, which in turn drives the logic circuit at least indirectly, in particular via a digital processor or DSP. This allows a very accurate and very fast control of the power supplied to a plasma process can be realized.

The total power generated and / or a power reflected by a load, in particular a plasma chamber, can be detected and fed to a higher-level controller. In this way, the output power of the power converter can be controlled or regulated.

Further features and advantages of the invention will become apparent from the following description of an embodiment of the invention, with reference to the figures of the drawing, which show details essential to the invention and from the claims. The individual characteristics can each individually for themselves or to several in any combination in a variant of the invention be realized.

A preferred embodiment of the invention is shown schematically in the drawing and will be explained below with reference to the figures of the drawing. Show it:

1 in a highly schematic representation of a plasma system with a power supply system;

2 in a block diagram, a power supply system;

3 a DDS module.

The 1 shows a plasma system 1 which is a power supply system 2 includes. The power supply system 2 again has a power converter 3 on, connected to a power supply network 4 can be connected. The at the output of the power converter 3 Power generated is via an impedance matching network 5 to a plasma chamber 6 given as a load, where a plasma is generated, by means of which a plasma treatment in the plasma chamber 6 can be carried out.

In particular, a workpiece can be etched or a layer of material can be applied to a substrate.

The 2 shows in a highly schematic representation of a power supply system 20 , The power supply system 20 has a power converter 30 which generates an output power that can be supplied to a load such as a plasma process or a laser excitation. In the power converter 30 are multiple amplifier paths 31 - 36 intended. The amplifier paths 31 - 36 are largely identical. Therefore, only the amplifier path will follow 31 described. The amplifier paths 31 - 36 have several amplifiers 37 . 38 which are suitable for amplifying an analog signal. At the end of the amplifier paths 31 - 36 is each an amplifier 39 provided with at least one LDMOS transistor whose output directly or indiekt, z. B. via an impedance adapter and / or filter, on a combiner 40 is switched. In particular, all outputs are all amplifier paths 31 - 36 , in particular in the same way, on the combiner 40 connected. By the combiner 40 become the individual powers of the amplifier paths 31 - 36 coupled to a total performance.

That the amplifier paths 31 - 36 are constructed identically, does not necessarily mean that they have the same gain. Component dispersion and circuit design tolerances can result in significant differences in phase and / or amplitude in the amplifier paths 31 - 36 generated high frequency power signals with identical input signal.

The amplifier paths 31 - 36 is each a DAC 41 upstream, which is a logic circuit unit 42 assigned. In particular, the DAC 41 from the logic circuit unit 42 Sequences supplied by digital values that make up the DAC 41 generates an analog output signal, which is an amplifier path 31 - if necessary after filtering through an optional filter 55 - is supplied. The DAC 41 and the logic circuit unit 42 can be used in a so-called direct digital synthesis (DDS) device 43 be integrated, also called: Direct Digital Synthesizer. Each of the amplifier paths 31 - 36 is a separate DDS building block 43 and accordingly a DAC 41 and a logic circuit unit 42 assigned. An example is the DDS module 43 in 3 described.

• 1. einen Signal-Datenspeicher 61, in dem Signal-Datenwerte zur Erzeugung der Analogsignalform abgelegt sind,
• 2. einen Amplituden-Datenspeicher 62, in dem Amplituden-Datenwerte zur Beeinflussung der Amplituden der Analogsignale abgelegt sind,
• 3. einen Multiplikator 63 zur Multiplikation der Signal-Datenwerte mit den Amplituden-Datenwerten und
• 4. einen Zähler 64, der dafür sorgt, dass in einem vorbestimmten Takt Signal-Datenwerte aus dem Signal-Datenspeicher 61 ausgelesen und dem Multiplikator zugeführt werden.

The logic circuit unit 42 indicates here:

• 1. a signal data memory 61 in which signal data values for generation of the analog signal form are stored,
• 2. an amplitude data memory 62 in which amplitude data values for influencing the amplitudes of the analog signals are stored,
• 3. a multiplier 63 for multiplying the signal data values by the amplitude data values and
• 4. a counter 64 which ensures that in a predetermined clock signal data values from the signal data memory 61 be read out and fed to the multiplier.

Both the signal data memory 61 as well as the amplitude data memory 62 can be designed as so-called look-up tables (look-up tables, LUT).

The DAC 41 also has a reference signal input 44 on which a drive circuit 45 may be upstream to generate a reference signal. The drive circuit 45 can be described in the Programmable Logic Device (PLD) to be described later 46 will be realized. The digital logic circuit may be implemented as a field programmable (logic) gate array Field Programmable Gate Array (FPGA).

By the at the reference signal input 44 input reference signal can be the output signal, ie the generated analog signal of the DAC 41 , be influenced, in particular amplitude modulated, be. The DDS building block 43 is the digital logic circuit 46 upstream, which also designed as a look-up table (LUT) memory 47 having. In the store 47 Amplitude data values may be stored by the memory 47 in the amplitude data memory 62 can be written. Also via the choice of the amplitude data values, an amplitude modulation can be effected. In addition, data for phase correction can also be stored. The values in memory 47 are used to compensate for differences in the amplifier paths ( 31-36 ) or in downstream combiners 40 , You can use the logic circuit unit 42 , in particular the amplitude data memory 62 be handed over. The digital logic circuit 46 is powered by a digital processor, especially Digital Signal Processor (DSP) 48 , driven, with a control panel 49 via a data bus 50 communicates.

The digital processor, in particular Digital Signal Processor (DSP) 48 , the memory 47 and the logic circuit units 42 can in a logic device 58 will be realized. The logic component can be used as a digital logic circuit PLD, in particular FPGA 58 be educated. Even if the DAC 41 integrated, results in a compact logic device 57 , The digital processor in particular DSP 48 , the memory 47 , the DDS building blocks 43 and also the DAC 41 as well as the filters 55 and the first amplifiers 37 can on a circuit board 56 will be realized. The identical circuit board 56 Can be used for a variety of different power supply systems with different power, different frequencies and different loads (laser, plasma, etc.).

The output power of the combiner 40 is via a directional coupler 51 trained measuring device to a load, eg. As a plasma process given. Through the directional coupler 51 the output power and a power reflected by the load can be detected. The measurement signal conditioning is done with measuring equipment 52 connected to the directional coupler 51 keep in touch. The measuring equipment 52 are in turn with the system control 49 connected. Thus, via the Control Panel 49 determine the output power of the combiner based on the detected output power and the detected reflected power 40 should be generated. According to this specification, the system control 49 the DSP 48 and this in turn the digital logic circuit 46 drive.

In the store 47 Digital values are stored which contain both amplitude information and phase and possibly also frequency information, so that at the output of the DAC 41 an analog signal with a predetermined amplitude, frequency and phase can be generated. This makes it possible, the output signals of the individual amplifier paths 31 - 36 to match, so this in the combiner 40 In particular, very simple combiners can be used in this way 40 Without loss-generating compensation impedances are used and a power loss can be kept low.

The inventive way to generate an analog signal, the power in the output of the combiner can very quickly 40 be influenced and changed.

Claims (7)

Method for generating a high-frequency power that can be supplied to a load by passing an analog signal through a DAC ( 41 ) and in an amplifier path ( 31 - 36 ), wherein the amplitude of the analog signal is modulated, characterized in that the amplitude of the analog signal is modulated by a digital signal input of the DAC ( 41 ) a series of digital values is supplied from which the DAC ( 41 ) generates the analog signal and generates the sequence of digital values by storing a signal in a signal data memory ( 61 ) stored signal data value with one in an amplitude data memory ( 62 ) multiplied amplitude data value is multiplied. A method according to claim 1, characterized in that the amplitude of the analog signal is modulated by the amplitude in the data memory ( 62 ) stored amplitude data value is changed. Method according to one of claims 1 or 2, characterized in that the amplitude of the analog signal is modulated by the in the signal data memory ( 61 ) stored signal data value is changed. Method according to one of the preceding claims, characterized in that the amplitude of the analog signal is modulated by a reference signal input ( 44 ) of the DAC ( 41 ) is supplied to influence the analog signal, a reference signal. Method according to one of the preceding claims, characterized in that in each case an analog signal (by a plurality of DAC 41 ) is generated and in an associated amplifier path ( 31 - 36 ) is strengthened. Power supply system ( 2 . 20 ) with a power converter ( 3 . 30 ) connectable to the load for supplying a load with power, the power converter ( 3 . 30 ) a first amplifier path ( 31 - 36 ), wherein the first amplifier path ( 31 - 36 ) one from a DAC ( 41 ) is supplied supplied amplitude modulated analog signal, characterized in that the DAC ( 41 ) a logic circuit unit ( 42 ) for generating a DAC ( 41 ) is supplied upstream of the supplied digital signal, wherein the logic circuit unit ( 42 ) a signal data memory ( 61 ) in which signal data values are stored to produce an analog waveform, an amplitude data memory ( 62 ) in which amplitude data values for influencing the amplitudes of the analog signals are stored, and a multiplier ( 63 ) for multiplying the signal data values by the amplitude data values. Power supply system according to claim 6, characterized in that the digital-to-analog converter ( 41 ) a reference signal input ( 44 ) and the reference signal input ( 44 ) a drive circuit ( 45 ) is connected upstream.

Images